magic-folder test failure on slackware buildslave (test_move_tree) #2834
Labels
No Label
0.2.0
0.3.0
0.4.0
0.5.0
0.5.1
0.6.0
0.6.1
0.7.0
0.8.0
0.9.0
1.0.0
1.1.0
1.10.0
1.10.1
1.10.2
1.10a2
1.11.0
1.12.0
1.12.1
1.13.0
1.14.0
1.15.0
1.15.1
1.2.0
1.3.0
1.4.1
1.5.0
1.6.0
1.6.1
1.7.0
1.7.1
1.7β
1.8.0
1.8.1
1.8.2
1.8.3
1.8β
1.9.0
1.9.0-s3branch
1.9.0a1
1.9.0a2
1.9.0b1
1.9.1
1.9.2
1.9.2a1
LeastAuthority.com automation
blocker
cannot reproduce
cloud-branch
code
code-dirnodes
code-encoding
code-frontend
code-frontend-cli
code-frontend-ftp-sftp
code-frontend-magic-folder
code-frontend-web
code-mutable
code-network
code-nodeadmin
code-peerselection
code-storage
contrib
critical
defect
dev-infrastructure
documentation
duplicate
enhancement
fixed
invalid
major
minor
n/a
normal
operational
packaging
somebody else's problem
supercritical
task
trivial
unknown
was already fixed
website
wontfix
worksforme
No Milestone
No Assignees
2 Participants
Notifications
Due Date
No due date set.
Reference: tahoe-lafs/trac-2024-07-25#2834
Loading…
Reference in New Issue
No description provided.
Delete Branch "%!s(<nil>)"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
5.g.: https://tahoe-lafs.org/buildbot-tahoe-lafs/builders/Markus%20slackware64%20stable/builds/279/steps/tox/logs/problems
This had been masked by #2807, which is now fixed. It's only happening on slackware, which is odd (travis and the other buildslaves are fine).
it seems this test is failing because inotify is behaving differently. a directory is moved into the watched directory so the new directory is watched now. the directory is moved out of the watched directory and then a file is added to it... the test expects the event to be captured... although this is the behavior we expect on linux... it's OK if a given inotify implementation doesn't work that way. it's not indicative of a bug in our software.
actually i'm convinced the test should always fail as it's written now because the last thing test_move_tree is testing is weather our inotifier watcher is watching a sub-directory which was moved into the magic-folder and then move out.
created a ticket to address that bug:
https://tahoe-lafs.org/trac/tahoe-lafs/ticket/2836
I patched the magic folder code to print out log messages, and forced a build of that branch.
The Jessie and Xenial builders, which both succeed, have the same pattern of events: "
local_dir/empty_tr\xc3\xaae
moved_to, is_dir" once, as expected.The Centos, Fedora, and Slackware builders, which fail, have a different pattern: "
local_dir/empty_tree
moved_to, is_dir" twice, then "local_dir/empty_tree
delete_self".It's unclear whether the filename being Unicode on the succeeding systems and ASCII on the failing systems is causative or a coincidence, given that the pattern of events is also different.
i modified magic-folder class and queue mixin to allow for debug logging after the magic-folder object has been created... that way in our tests we can easily turn debug logging on for a particular test as i've done here for test_move_tree:
https://github.com/david415/tahoe-lafs/tree/2834.logging_test_move_tree.0
however this change uses 'print' to log the messages instead of twisted logger. i suppose we cannot merge this until that's fixed.
yeah, make it a twisted.python.log.msg() and they'll show up in test.log where the buildbot can show them to us
In 250c6f60/trunk:
this TODO and this ticket can go away once we've landed our new inotify test suite from #1432.
it will improve reliability of tests by only making a small set of inotify tests actually test the inotify implementation. all the magic-folder tests will use a mock inotify.